Controlling Births and Bodies in Village China

by Susan Greenhalgh
Citation
Title:
Controlling Births and Bodies in Village China
Author:
Susan Greenhalgh
Year: 
1994
Publication: 
American Ethnologist
Volume: 
21
Issue: 
1
Start Page: 
3
End Page: 
30
Publisher: 
Language: 
English
URL: 
Select license: 
Select License
DOI: 
PMID: 
ISSN: 
Abstract:

controlling births and bodies in village China

SUSAN GREENHALGH-University of California, /nine, and Research Division, The Population Council, New York

From the intrauterine device (IUD) to the pill to the subdermal implant, modern birth control methods have been both emancipating and shackling for women. While providing ever more efficient means of avoiding unwanted pregnancies, these technologies have placed women and their bodies under the control of powerful state and medical establishments.

By world-historical standards, China's birth control program has been exceptional in its hostility to women. It is women's bodies that have been made to bear the burden of contracep- tion and abortion, and women's private and public selves that have been diminished by the policy's prescriptions and social sequelae.

Understandably, most observers have viewed Chinese women as victims of the one-child birth program. But a closer look at the micropolitics of policy implementation suggests that women are not only victims, but also agents in the practice of controlling births and making informal population policy in China's villages. Field research in the northwestern province of Shaanxi shows how, by resisting some of its harshest provisions, Chinese women, acting for themselves and their patriarchal families, have shaped thedevelopment of village fertility policy and, in turn, its effects on their bodies, reproductive outcomes, and options for living.

In this article I draw on three fields of anthropological inquiry--one feminist, another demographic, a third politicaleconomic-which I hope to bring closer together.
feminism, family demography, and political economy: perspectives on reproductive politics

Since the birth of "second wave" feminism in the late 1960s,' the politics of reproduction has been a central concern of feminist thought and practice (an excellent review of this work is Ginsburg and Rapp 1991 ). Historians of Western societies have documented the progressive

This article seeks to deepen the understanding of reproductive politics by conjoin- ing a feminist analytics of reproductive control with a demographic dissection of reproductive process and outcome, as well as a political-economic enquiry into state domination and accommodation. Focusing on China's one-child-per-family birth control program, it argues that women are not only victims but also agents in the practice of controlling births and making population policy in China's villages. In Shaanxi Province, peasants have contested policy elements they do not like, forcing local officials to negotiate the terms of policy implementation. Resistance to the policy has had contradictory effects, however: while increasing the number of children allowed, it has put women's bodies at risk and reinforced their social subordination. Ironically, resistance has worked to reproduce the very state control over childbearing that women have contested. [reproductive politics, the body, feminism, demography, political economy, China]

Arneriran Ethnologrst 21(1):3-30. Copyright O 1994, American Anthropological Association.

controlling births and bodies

loss of control over reproduction experienced by women since the 19th century, when childbirth began to be subject to growing medical oversight (Ehrenreich and English 1978; Gordon 1977; Oakley 1984; also Donzelot 1979; Foucault 1978). Students of contemporary society have underscored the contradictory effects on women of developments in reproductive technologies, including those for fertility control and the management of childbirth, ante- and neonatal care, and infertility (Rapp 1990; Rowland 1987; Stanworth 1987). Medical and scientific advances in these areas have clearly expanded women's ability to decide if, when, and under what conditions to mother. At the same time, however, because access to and use of these technologies tend to be dominated by the medical profession, their development has increased the potential for others to control women's lives.

In third world societies, where most anthropologists work, the locus of institutional control over reproduction tends to be the organized family planning program. Critics of these programs charge that most consider demographic objectives as paramount, losing sight of, at times even abandoning, concerns for women's health, safety, and freedom of reproductive choice (Bruce 1987; Hartmann 1987; Warwick 1 982).2 Today most programs have a range of contraceptives available, but few have an underlying ethic or offer provider training that supports client ~hoice.~ In practice, most programs promote the use of a handful of surgical or systemic female methods, in particular, the IUD, the pill, and tubal ligation. Despite their greater health and safety risks compared to those of barrier methods such as the diaphragm, cervical cap, and condom (Bruce and Schearer 1983; Dixon-Muel ler and Wasserheit 1991 ), these methods are generally pre- ferred because they are highly effective and, in the case of the IUD and sterilization, because they place control in the hands of program personnel (Hartmann 1987).

Feminist anthropologists have shown that women generally are not passive victims of efforts by others to control their reproductive experiences. Working primarily in the United States, where the medicalization of reproductive health care is extreme, these scholars have demon- strated that women are active creators of their own reproductive outcomes, accepting images, technologies, and practices that serve their interests while rejecting those that do not. For example, Martin has shown how the medical model of birth, with its metaphors of machines and mechanics, demeans women and how women resist that model by using it in subversive ways and creating alternative birth imageries that accord them more dignity and autonomy (Martin 1987, 1990, 1991 ).

Individual and group protests serve not simply to contest malpractices but also to alter them. The process is depicted as one of negotiation, in which women, despite unequal resources and power, contest the denigrating or even harmful practices of more powerful actors, and at times transform dominant ideas and practices to better serve their interests (Ginsburg and Tsing 1990; also Ginsburg 1989). One must not romanticize resistance, h~wever,~

for its effects are often contradictory. Depending on how, when, and by whom they are waged, struggles over reproductive control may end up empowering women, perpetuating the status quo, exacerbat- ing existing inequalities, or some combination of the three (see, for example, Rapp 1990,1991 ). Furthermore, in Asia, Africa, and Latin America, women's resistance to the encroachment of formal institutions on the domain of childbearing may express not only their own predilections, but also those of the patriarchal families to which they belong.

Feminist scholars have been primarily concerned with the effects of reproductive micropoli- tics on bodily practices regarding birth and the cultural construction of gender. But work in family demography suggests that struggles and negotiations underlie not just the processes of pregnancy and childbirth, but also the outcomes, including the number of births a woman has and, more important from an anthropological perspective, the number and sex of the children she treats as part of her social family. The bulk of this work has been done in the third world, where "overpopulation" and its cure, "the demographic transition," are subjects of intense academic and political debate.

american ethnologist

Challenging the mainstream demographic view of fertility as a one-time biological event (childbirth), anthropological students of family demography have reconstrued reproduction as an ongoing social and political construction that may begin long before and continue long after the biological fact of parturition (a clear and cogent statement is Bledsoe 1990). As a social (and cultural) construction, reproduction involves the manipulation of a wide range of cultural resources and social institutions, from medical secrets (Browner and Perdue 1988), social labels (Launay, in press), and child care practices (Levine 1987) to marriage (Fricke, in press), fosterage (Bledsoe and Isiugo-Abanihe 1989), and adoption (Wolf and Huang 1980). Reproduction is a political construction because the strategies deployed--concealment, deception, collusion, collaboration, and so forth--entail the negotiation of power relations with husbands, co-wives, lineage elders, and other members of one's social network.

While feminist and demographic students of reproduction within anthropology have devel- oped separate, largely noncommunicating discourses, joining them might be advantageous. Intellectually, the two subfieldsof inquiry share many of the same theoretical roots.5 Both reflect anthropology's current concern with such issues as human agency (Giddens 1979), the social construction of reality (Berger and Luckmann 19661, and power as a structuring principle of social life (Roseberry 1989). The addition of a feminist analytics of control to the demographic dissection of reproductive outcomes allows us to broaden the domain of political inquiry by bringing the body as a locus of reproductive struggle and negotiation into the picture. Carter's (in press) highly original essay on "the body resourceful" suggests how such a feminist demography might look.

To complete the picture of third world reproductive politics, the insights of another subfield, political economy, are required. Political economy lifts our sight above the local system to the national and global arenas, where state bureaucracies, international organizations, and multi- national corporations conduct their business (Roseberry 1988; Wolf 1982; on demographic political economy, Greenhalgh 1990a). While our knowledge of the demographic impact of these demographically powerful actors remains sketchy, historical European demography indicates that states have substantial indirect effects on fertility that flow from their policies on child labor, school attendance, and other aspects of social and economic life (Kertzer and Hogan 1989; also Schneider and Schneider 1984).

In contemporary third world societies, the state also affects family formation directly through itsefforts at fertility control. Often supported by foreign governments and international agencies, roughly three-quarters of developing countries with populations of one million or more currently have family planning programs (Ross et al. 1992:80-81). By definition and design, these programs impose themselves on women's bodies and families' reproductive deliberations. By promoting modern contraception and small-family norms, state policymakers and program personnel become key parties to the reproductive negotiations that go on at the microlevel. These people too are agentsand usually powerful ones at that-whose goals and tactics need to be brought into the analysis of reproductive politics.

A good place to conjoin the perspectives of feminism, demography, and political economy is the People's Republic of China. In that country, key issues within all three perspectives are both highly salient and intensely contested aspects of reproductive life. Especially since 1979, when the one-child policy was introduced, the party-state has sought to strengthen its grip on reproductive behavior. At the same time, it has seized firm control of the issue of family size, one of the few matters Chinese families had managed to decide for themselves during most of the Maoist era.

China is of special interest too because it shares pride of place with Ceausescu-era Romania, in having subjected reproduction to what may be the world's harshest political controls. In such places, reproductive autonomy is a largely unattainable goal. People's aims here are much more

controlling births and bodies

concrete: to securethe long-term survival of the family and to minimize the bodily harm suffered by women. The stakes in this contest are very high to state, family, and individual women.

While presenting unique opportunities, China poses special problems for anthropological research on reproduction. These problems arise from the constraints that continue to be placed on fieldwork by foreigners6 and, more important, from the political sensitivity of the birth control program in both domestic and international arenas. These political facts of life somewhat narrowed the range of information I was able to gather in my research. Because the project was completed before the Tiananmen clash of June 1989, however, I was able to collect a large amount of information which, I hope to show, sheds new and important light on reproductive politics in that little-understood country.

Interweaving elements of the three approaches sketched above, I develop a "negotiational" approach to the "production of reproduction" in China's one-child birth control program. Using field data from a village in the northwestern province of Shaanxi, I examine three aspects of reproductive micropolitics and their embeddedness in the changing political and economic structures of village life: resistance to the birth control program; negotiation over family size and contraceptive practice; and the consequences, both beneficial and deleterious, for women, their bodies, and reproductive outcomes.

Before launching into such an enterprise, one must ask what concepts such as resistance and negotiation mean in the political context of the People's Republic. Recent research on state-society relations makes clear that these notions are meaningful even in China, where the party-state has sought to dominate virtually every aspect of social life. As Shue (1 988) has argued, even in the Maoist era the cellular structure of rural society enabled local cadres to protect peasants from the extremes of state power, providing scope for resistance to and negotiation of unpopular policies sent down from above. The room for local opposition has greatly expanded in the reform era, when the dismantling of the collectives reduced cadre control over the peasantry (see Burns 198546; Unger 198546; Oi 1989 takes a more cautious view). One must not conclude, however, that the party-state is absent from peasant life today. In her political ethnography of a South China locality, Siu maintains that state power has been so strong over the years that people have come to internalize it, to accept it, and even to reproduce it in their daily lives (1 989:ll). As we shall see, this argument finds ample illustration in the area of birth control.
birth planning in China

In 1979, when the one-child policy was announced, the concept of "birth planning" was already 25 years old. An indigenous Chinese-indeed, Maoist-strategy (White 1992), birth planning (jihua shengyu) differs from the Western liberal notion of family planning in that the role of the party-state (heweforth simply "state") is paramount: births are planned by the state to bring the production of human beings in line with the production of material goods.

the one-child policy and program The one-child policy followed on the heels of two small-scale campaigns, conducted in 1956-58 and 1962-66, and the nationwide "later-longer- fewer" (wan xi shao) policy of 1971 -78 (for details see Chen and Kols 1982 or Banister 1987). Under the later-longer-fewer policy (so named because it encouraged later marriage, longer child spacing, and fewer children), peasant couples were allowed to have three children and then, beginning in 1977, two. During the 1970s, fertility fell from six to just under three children per woman (Coale 19841, an extraordinary decline by any standard. Nevertheless, the post-Mao leadership became convinced that fertility had to fall even further-to about 1.7; some even hoped for only one child per woman (Liu 1981 ;Song 1981 &if Chinese society was to become

american ethnologist

"comfortably well off" (xiaokangshehu~] by the turn of the century and the regimewas to secure its legitimacy.

Largely unknown to outsiders, the one-child policy underwent substantial modification during its first decade. The policy that in January 1979 sought simply to encourage single-child families was soon abandoned and in September 1980 replaced by the call for all couples to limit themselves to one child. The coercive phase of the early 1980s was capped by a massive nationwide sterilization campaign conducted on and off throughout 1983. In part because of widespread resistance to the campaign-in which some 20 million people were sterilized (Ministry of Public Health 1988:79&the policy was relaxed somewhat in the mid-1 980s. Rules on second children were liberalized slightly in mid-1 984 (Greenhalgh 19861, and in 1988 rural couples whose first child was a girl were permitted to have a second child (Zeng 1989). Throughout the decade, third births continued to be "resolutely prohibited" (jianjue dujue). During these years the policy in Shaanxi Province and Xianyang City, where the villages I worked in are located, followed national policy fairly closely. In the area studied, the number of conditions under which peasant couples were allowed second children rose from three in 1981 to sixteen in 1985,' and in 1986 villagers whose first child was a girl were permitted to have another (for details see Greenhalgh 1990~).

Crucial for women and their bodies, birth planning is target driven. Population growth targets are devised at the political center and then handed down, level by administrative level, until they reach the point of implementation. There, township and village birth-planning cadres are expected to place the achievement of state-assigned goals above concern for the self-defined reproductive desires of individual women and families.

Clearly, the issue of individual reproductive rights stressed by Western feminists has no place in official Chinese discourse. China takes the position that individual childbearing is a matter to be decided by the state for the good of society as a whole. The masses are to be persuaded that what is good for society is good for them as individuals. For married couples, birth control is mandatory, a constitutional obligation.' Thus, the study of reproductive negotiations between state policymakers and local society would appear to be unrewarding in the Chinese context.

If one looks beyond official discourse to unofficial practice, however, another universe of expression opens up. The state may want to control reproduction, but groups within society have contested state domination of this crucial area of social life. In villages and byways throughout the country, peasant families have devised myriad means of evading state control so as to get the number and sex of children they deem best for themselves (evidence from the Chinese press is presented in Wasserstrom 1984). By pressuring local policy enforcers to relax the rules, they have promoted the evolution of an "informal policy" that differs from the formal policy embodied in the written regulations. Thus, in China the extent and forms of resistance to state restraints on reproduction, and the effects of that resistance on informal birth policy and on women's bodies and reproductive outcomes, are fruitful issues for investigation.

birth control: the control of women's bodies Promoting birth control has been a major- since 1981 the major-means of controlling fertility.9 While government propaganda has stressed the importance of a "cafeteria" approach to contraception, the selection of methods made available to the rural population has been narrow and focused on two low-cost, highly effective, provider-implemented procedures-IUD insertion and sterilization-that involve only one-time motivation on the part of the user and little or no attention to use from that time on. As long-term, and in the latter case permanent, methods requiring bodily intervention, these techniques place control over contraceptive practice in the hands of state policy enforcers.

Surgical and other provider-controlled means of contraception totally dominated the birth control program during the 1980s, when the emphasis on achieving targets made them indispensable tools in the implementation kits of local cadres. The highly precise family-size

controlling births and bodies

rules of the one-child-with-exceptions policy also fostered increased use of abortion, which was a handy if unpleasant means to eliminate pregnancies of unauthorized parity. In fact, in the 1980s birth control was virtually synonymous with what the Chinese call "the four operations" (sizhong shoushu): IUD insertion, tubectomy, vasectomy, and induced abortion. Statistics compiled by the Ministry of Public Health indicate that between 1979 and 1987 an astonishing 269 million such operations were performed nationwide (Ministry of Public Health 1988:79).

Given the country's vast population and scarce resources, emphasis on surgical and other provider-implemented techniques was a reasonable strategy from both public health and economic points ofview. However, thecontrol aspectsofthis strategy should not beoverlooked. There can be little doubt that removal of control over contraception from the masses was a deliberate objective of Chinese population policymakers. A few historical facts must suffice to make the point. First, the supply of oral pills, the one highly effective means of contraception that allows an individual to change her mind without getting state permission, has been limited in rural areas (Chen and Kols 1982:)592). In fact, as the birth control program operated in most places, including the one I studied, pills were available only to women physically unable to carry an IUD, the choice being made not by the women themselves but by authorized, not always medically competent, birth cadres (the situation was similar in Fujian; see Huang 1 989:181). Second, the IUDs developed in China were deliberately designed to be tamperproof. Until the early 1990s, the most commonly used IUD was stainless steel and had no string, making it difficult for thewearerto check its location and physically dangerous for hertoattempt nonclinical removal (Banister 1987:206). The state's desire to promote reliance on "the four operations" can also be read from its massive efforts to facilitate their use in every village in the country. Barefoot doctors were trained to perform IUD insertions, early abortions, and sterili- zations, and special equipment was designed for use by the mobile medical teams that took birth control operations to villagers in remote areas (Chen and Kols 1982:J594).

Individual control over contraceptive decisions was greatly narrowed in late 1982, when, in preparation for the 1983 sterilization campaign, the state mandated IUD insertion for all women with one child and sterilization (male or female) for couples with two or more. Local campaign documents make clear that these rules were in effect for the duration of the 1983 campaign in the villages I worked in." During other times the provincial and municipal regulations guiding birth control in the villages instructed that all couples of reproductive age must adopt effective contraception, and advocated that, with a few specific exceptions, women with one child use an IUD and one member of couples with two or more be sterilized." "Remedial measures" (bujiu cuosh~)-the official euphemism for abortions-were required for all unauthorized pregnancies.

Birth control meant controlling not just any bodies, but, more particularly, women's bodies. Despite China's status as a pioneer in research on male methods, in China, as elsewhere, the great majority ofcheap and reliable methods are designed for use by women. Even sterilization has been performed largely on women. The government tried hard to promote the simpler male procedure, vasectomy, especially in the 1960s, but strong resistance from men worried about its effects on their bodies and libido has meant that the great majority of sterilizations have been the more costly and, where medical facilities are poor, somewhat more dangerous tubal ligation (for more on men's concerns see Huang 1989:181 ).I2

Women's bodies were not only vehicles for the enforcement of the birth policy; they were also sites of struggle for control over reproduction. Before turning to the politics of birth control in the area studied, we take a look at the politics surrounding the research, and its effects on the information gathered.

american ethnologist
the production of demographic knowledge

We know that social science research is an inherently political process in which the power differentials between researcher and researched have marked effects on the information gathered (Rosaldo 1989). My project, which focused on a highly charged issue, exhibited these problems to an extreme degree.

Designed to illuminate the political economy of reproduction in three villages over a 40-year period, the research could not have been carried out without the full cooperation of Chinese scholars and local officials. The data were gathered as part of a field project conducted in collaboration with Xi'an Jiaotong University's Population Research Institute. (The interpretations offered in this article, however, are solely my own.) Carried out during the first six months of 1988, the project involved a demographic survey of all ever-married women in the villages (1,011 women, 409 in the largest village, which is the focus here), the gathering of social and economic histories from 150 families, in-depth interviews with present and former officials (known in China as cadres), and documentary research in local newspapers, journals, and other sources. During the research period, I lived in the largest village for one month, commuting from Xi'an the other five. While all interviews were conducted in the privacy of people's homes or family fields, with no cadre present, the project was formally sponsored by the township government; indeed, our local sponsor was none other than the deputy township head, one of whose duties was to enforce the birth-planning policy.

This was a situation fraught with potential for the production of politicized knowledge. The villagers, who were largely powerless in these interactions, sought to satisfy their curiosity about me and my scholarly collaborators without revealing anything that would subject them to further controls on their private lives. My co-researchers were clearly part of the dominant discourse and practice. Employed in a prestigious, state-run university, their role was not to produce independent evaluations of the merit of fertility regulation, but to assist the state in its policy efforts.13 Township and village birth-planning officials, I believe, hoped to use our research to demonstrate their own success in controlling reproduction in the villages. For my part, I endeavored to elicit evidence of discontent and resistance to the policy without jeopardizing the larger data collection project. Thus, I sought to subvert dominant practice even as I helped to uphold it.

What were the effects of these varying objectives and positions on the information gathered? Because I was largely constrained from asking direct questions about resistance, the formal record of field notes, interviewtranscripts, and questionnaire data contains few overt challenges to state policy. The formal record certainly contains no information that would enable me to write a truly cultural account of birth planning, replete with emotional detail. Other parts of the record, however, tell a different and potentially useful story. In her fascinating work on women and language, Gal (1 991 ) has shown how in certain power-laden verbal encounters, including the ethnographic interview, silence can serve as a means of resistance, a political protest against explicit or implicit domination. I believe that in their conversations with us, both peasants and cadres made strategic use of silence to protest aspects of the policy they did not like. Cadres, for example, were loathe to comment on birth-planning campaigns; peasant women were reluctant to talk about sterilization. These silences form one part of the unofficial record of birth planning in the villages. More explicit protests were registered in informal conversations. From these interactions emerged a sense ofthe profound distress of villagers forced to choose between a resistance that was politically risky and a compliance that violated the norms of Chinese culture and of practical reason.

A third source of unofficial information is the official survey questionnaire. Village women could not openly relate the stories of their resistance, but they could reveal them-in fact, had no choice but to reveal them-in the contraceptive and fertility histories that were collected in

controlling births and bodies

the formal interviews. Read closely, these histories chronicle cadre strategies, peasant resis- tances, and cadre counterstrategies in temporally exact if emotionally "muted" (Ardener 1975) detail. These histories speak volumes about the negotiation over birth control in the villages. Informed by the casual conversations and nonresponses, these reproductive histories, along with individual cases, will form the basis for my discussion below.
the village setting

Located in the Wei River Basin just west ofthe Shaanxi provincial capital of Xi'an, the villages studied are part of Weinan Township (a pseudonym; 1987 population 26,102), which in turn is part of Xianyang City (1 986 population 3.97 million). This article focuses on the largest of the three villages, which I call Shijiacun (Shi Family Village).

social and economic life in Shijiacun In 1987 Shijiacun was a Han village of 1,623 people living in 362 registered family units. The village was not always so large. In 1958, on the eve of the Great Leap Forward, only 842 people lived within its borders. In a story that could be told of countless villages around the country, however, high fertility in the 1950s and 1960s produced rapid population growth, and, in turn, intensifying pressure on ecological and economic resources. It is these pressures policymakers sought to ease when they fashioned the one-child policy in the late 1970s.

Vegetable growers for the nearby urban population, the villagers were moderately we1 I-to-do by Shaanxi standards, although only lower-middling by Chinawide criteria. In 1987, per capita income in the township (village income data are not available) stood at 122 percent of the provincial average (for the agricultural population), but only 87 percent of the national average.14

The township's modest level of prosperity was largely a product of the economic reforms of the late 1970s and early 1980s. After inching up from 66 yuan in 1962, the first year of the post-Great Leap consolidation, to 11 5 in 1980, per capita income soared to 402 (about 325 when adjusted for inflation) yuan during the first seven years of the reform decade.

Rising incomes brought a rash of new home construction and acquisition of consumer goods. When I was in the village in 1988 two out of five families had a television, and many boasted washing machines, motorcycles, and other trappings of modern life. Although poorer villagers continued to live in the yellow mud-brick houses characteristic of the loess plateau, richer ones had constructed two-story concrete homes spacious enough to accommodate one or more married sons. Walled courtyards continued to protect the privacy of old-and new-style homes. In the evening, family life revolved around the kang, the large built-in bed warmed by smoldering corn husks placed in a heat chamber below.

Completed by early 1983, thedismantlingof thecollectives and distribution of land to peasant househ~lds brought pronounced shifts in the units structuring social life. The production brigade-turned-village (technically, villagers' committee) retained some, but diminished, influ- ence over peasant activity. Village-level cadres possessed a certain amount of leverage based on their control over responsibi lity land, housing plots, chemical fertilizer, and electricity. Cadre power was undoubtedly weaker than in the typical grain-growing village, however; for, unlike grain growers, the vegetable producers I studied were not required to sign production contracts with the collective. In Shijiacun, the peasants enjoyed near-total control over the use of their plots and the marketing of their produce. As we shall see, these shifts in the balance of power expanded the scope for resistance to the birth control policy.

The village's seven production teams sustained much more damage. In the past, virtually all of economic life was organized by the teams. (Only tiny private plots were free from collective control.) With the introduction of the household responsibility system, the organization of

10 american ethnologist

production, marketing, and consumption reverted to the household, leaving the teams largely functionless. Although team-level positions continued to exist on paper, they were seen as conferring little power. Most team cadres were indistinguishable from their neighborsfull-time peasants who devoted all their energies to getting rich.

With collective units largely if not wholly peeled away, thefamily-household (the two largely coincided) emerged as the near-exclusive focus of peasant life. It was the unit of economic activities; of social responsibilities such as child care, socialization, and old-age security; and of cultural events including the weddings, funerals, and other life-cycle rituals that played a prominent role in village life. Decades of socialism-induced poverty had worn away much of the family sentiment that contributed to the development of extended families in the past. Generational conflicts were not far from the surface, leading to early-in cases extremely early-family divisions and, in consequence, a predominance of small nuclear units.''

Gender inequalities were in many ways deepened by the reforms. Men took the jobs outside agriculture, leaving their wives to tend the family farm. Certain lucrative activities were simply off bounds for women. For example, women, who were considered polluting, were not al lowed to work on the newly formed construction teams because it was believed the houses would collapse if they did. The enhanced valueofchild labor in the family farming regimeencouraged parents to take some of their children out of school early. Girls were removed from school as early as the fifth or sixth grade, while their brothers were encouraged to complete junior middle, and, if possible, go on to senior middle school. Differences in opportunities in adult life fed the traditional preference for sons, which was articulated ad nauseam in our interviews (on this, more below) and evidenced culturally in such customs as the one-month ceremony, which was more lavish if the baby was a boy.

the makings of a reproductive struggle The struggle over reproduction focused on two issues, family size and contraceptive use. Simplifying for analytic clarity, we can identify two main sets of views or norms on each issue, those of state policymakers and those of the villagers. State norms were reflected in the guidelines of formal policy, discussed above. This section describes the village norms.

Given its intent to shrink the most important resource available to Chinese peasant society- the family-it is not surprising that the birth control policy met tremendous hostility in villages throughout the country, including Shijiacun. For reasons more fully elaborated elsewhere (Greenhalgh 1993), the residents of Shijiacun believed in the late 1980s that their long-term well-being hinged crucially on their raising two children, at least one of whom was a son. More than two was acceptable, but two was best. The very best was one son and one daughter. A second son might substitute for the daughter, but a second daughter could never substitute for the son.

These beliefs about desirable family size and composition (these are convenient demographic shorthand for number and sex of children) were no mere survivals of what the state calls "feudal tradition," although cultural traditions, in particular the desire for a son to perpetuate the family line, certainly played a role in their formation. As abundant research on Chinese rural life has shown, however, these views were also firmly rooted in the exigencies ofvillage life, especially the need for labor and the absence of nonfamilial sources of old-age support (see, for example, Parish and Whyte 1978; Whyte and Gu 1987). Child mortality risks, while much lowerthan in the past, were not negligible either. The very real possibility that one child might die in an accident-vehicular deaths of children had risen frighteningly with the increase in road traffic in the 1980s-made it absolutely essential to have two children.

As virtually identical views on these matters were expressed by villagers of both sexes and all age groups, they functioned not only as individual family, but also as community reproduc- tive norms. Two such community norms can be distinguished: the village ideal was one of each

controlling births and bodies 11

sex. The village minimum-that deemed essential for social and economic survival-was two children, including one son.

Peasant views on family size were so uniform in part because of the molding power of two decades of forceful state policy on reproduction. The larger socioeconomic environment certainly supported small-family norms, but it is unlikely that the peasants would have declared only two children as ideal had the state not defined first three, then two, and finally one as the official norm. My argument is not that the villagers were cleverly or cynically concealing their "real views" from me. It is, rather, that their real views were so colored by decades of state propaganda and, more importantly, limited reproductive possibilities, that what was demographically thinkable bore the firm imprint of state desires for fertility limitation. By embodying state demands in their conscious reproductive aspirations, the peasants were not only accepting, they were also unwittingly reproducing state control over their childbearing.

Perhaps because little in their environment encouraged them to adopt official views on birth control, women's attitudes toward contraception ran counter to those of the state. While state policymakers advocated long-term effective contraception, women wanted to keep their options open and their bodies free of outside interference.I6 Some women, especially those in their thirties and forties, undoubtedly were glad to have their worries about becoming pregnant removed by the highly effective methods offered by the state program. In general, though, those I talked to were apprehensive about undergoing surgery of any type. Women in their fifties, long past the risk of pregnancy, continued to carry IUDs that had been inserted a decade or more earlier because they were afraid of the medical ordeal they would have to endure to have them removed. Village birth-planning cadres confirmed the existence of such fears, noting how much women dreaded sterilization in particular (similar fears have been voiced elsewhere; see Huang 1989:181; Potter and Potter 1990:247).

While evidence of the extent of medical problems is hard to come by, the Shaanxi and Xianyang birth-planning regulations have consistently stipulated ways to treat complications and accidents, and they have urged medical personnel to upgrade the quality of contraceptive operations. The very existence of such provisions suggests that medical problems, especially with sterilization, may be far from rare." The birth cadre in Shijiacun noted that infections have been a problem, especially during large sterilization campaigns. Although contraceptive use may not have been associated with widespread medical problems (the relevant information has not been gathered or made public), only a few serious cases need come to light in any given locality to arouse general worry.

Anxiety over possible physical complications was compounded by the fear of being targeted for an operation during a birth control campaign. Launched by central or provincial party leaders, campaigns, sometimes called "shock months" (tuji yue), involved short periods of one to several months of intensive mobilization of the population to achieve specific birth control targets. Huang's extraordinary description of a birth control campaign in Fujian reveals the anxiety and panic that may attend these affairs, especially when they are run by outside work teams who have little compunction about forcefully entering people's homes to find pregnant women and removing the valuables of those who refuse to comply (1 989:175-185). Because the program had relied heavily on campaigns over the years, birth control surgery had cometo be associated in the minds of people in Shijiacun with heavy pressure and rushed operations performed in sometimes inadequate facilities. The campaign-surgery link intensified the desire to avoid the operations at any cost.

Although individual women were on the front line of the birth control battle, it is important to recognize that they were acting as agents not only for themselves, but also for the patriarchal families to which they belonged. Indeed, as the Potters have argued, Chinese culture assigns reproductive rights not to the individual woman, but to the family into which she marries (1990:235). In Shijiacun, as in the Guangdong villages where the Potters worked, not only

12 american ethnologist

women but also their husbands and parents-in-law, and sometimes even members of their natal families, who had important roles in the rituals surrounding procreation (Huang 1990), participated in decisions on reproduction. On the issue of the ideal number ofchildren, there appeared to be little discord among family members. However, one can imagine that friction developed when families faced difficult choices. Although I do not have information on this subject, press reports from other parts of the country suggest the existence of clashes over such issues as what a couple whose first two children were girls would do. Thus, although women and their bodies were on the battle line in the struggle for reproductive control, just behind them were their husbands and other powerful family members, supporting, prodding, and, at times, probably pressuring them to resist.

As the bottom rungs of the state apparatus, township and village birth-planning cadres were obliged to enforce provincial and municipal policy as part of their official duties. As members of local society, however, village cadres were predisposed by cultural values and social obligations to soften state policies to accommodate pressing peasant demands (Shue 1988). (Township cadres were generally from outside the area; this was the case in Weinan.) The women's head and chief birth-planning cadre of Shijiacun, a woman of 39, had served in that capacity since 1969, the year she married into the village. From many late-night conversations with her (I stayed in her home), I believe that in her heart of hearts she held most of the views I have just described as belonging to ordinary peasants. So did the local communist party secretary, whose word on everything was final. The birth cadre sympathized with the need for two children and a son, and empathized with women who feared contraceptive surgery. In fact, she herself was not sterilized although she had three children (she did, however, use an IUD). In her talks with me, she expressed willingness to enforce the policy in a reasonable manner, bending the rules somewhat for people in difficult circumstances (such as having two daughters but no son). However, when birth control campaigns were launched at higher levels and actively promoted by city and township officials, she felt she had little choice but to do her best to reach the targets, as long as doing so did not require her to violate sacrosanct values.

the changing context of policy enforcement The scope for resistance to state policy--or, from state policymakers' point of view, for successful enforcement of state policy-has been crucially dependent on the organization of social, economic, and political life atthe village and township levels. The tremendous demographic success of the later-longer-fewer policy was due in no small measure to the control over peasant livelihoods wielded by cadres in the collective rural socioeconomy of the 1970s. The liberalizing economic and political reforms of the late 1970s and early 1980s loosened cadre control over the peasantry, undermining enforcement mechanisms and greatly expanding the room for resistance to unpopular policies such as that demanding one-chi Id families.

While this is not the place to elaborate on the devastation the reforms wreaked on enforce- ment instruments, a few of the main disruptions can be noted (for more see Greenhalgh 1993). The abolition of collective income accounting, combined with the rise in peasant incomes, disabled the system of economic incentives and disincentives, the key mechanism for enforcing the birth control policy. In Shijiacun, the system of fines for "unplanned births" began to totter in 1982 and collapsed completely in 1987. Political reforms, completed in 1984, exacerbated implementation difficulties by reducing the power and prestige of local cadres.'$ Village-level and team-level cadres were included in the distribution of collective land, depressing their commitment to the difficult and unremunerating work of population control. In Shijiacun, the main village-level birth-planning cadrecontinued to perform this work to the best of her ability, at times with the assistance of other village-level cadres whose main responsibilities lay elsewhere. However, she had little help from the team-level women's cadres, who found cultivating their fields infinitely more rewarding than badgering their neighbors to contracept.

controlling births and bodies 13

As if this were not enough, funds for birth control work began to dry up in the 1980s, complicating efforts to provide safe, timely operations.

Changes in villagedemography compounded the problemsfaced by birth cadres in Shijiacun. After increasing slowly from the 1950s through the 1970s, their target population-married women of childbearing age-ballooned in the 1980s. If it was difficult to manage the repro- duction of 200 or so women in 1979, the problems of coping with over 300 in 1987 were overwhelming. With few economic or administrative means left to enforce state policy, cadres had little choice but to accommodate peasant desires and negotiate new, locally acceptable terms of implementation. To see what those terms were, we begin by juxtaposing state policy with village practice on birth control.
state policy and village practice

As we have seen, state policy required all married couples to use contraception. In Shijiacun, contraceptive use increased rapidly during the 1970s, only to level off during the 1980s. During 1979-87, the proportion of married women of reproductive age (those aged 15 to 49, hereafter simply "women") who were contracepting hovered around 60 percent, rising slightly in some years and falling modestly in others (see Figure 1). In China as a whole, according to official figures, contraceptive levels in 1987 stood at 77 percent; in Shaanxi they reached 84 percent (China Population Information and Research Center 1990; Shaanxi Provincial Bureau of Statistics 1987:65). In Shijiacun, then, compliance with the policy was poor by comparative standards, if we can believe the official figuresa very big if.''

Trends in contraceptive methods used, also shown in Figure 1, reflect clearly the emphasis on surgical and other provider-controlled methods built into national and provincial policy. Client-controlled means, primarily oral pills, never accounted for more than 2 or 3 percent of contraceptive use in the village. Throughout the period, the vast majority of contracepting couples, about 70 to 80 percent, used the IUD to avert conception. This emphasis was reduced slightly in 1983, the year of the huge sterilization campaign. The proportion of contracepting couples sterilized rose from 18 percent in 1979 to 28 percent in 1983, and remained at or near that level through 1987. Fully 90 percent of these sterilizations were performed on women. Despite the rise in sterilization in the 1980s, there was no wholesale switch to the method, as we might expect if the policy of advocating (in 1983 insisting on) sterilization for virtually all couples with two or more children had been enforced with vigor. Clearly, the Shijiacun cadres settled for something short of full adherence to state rules on contraceptive type.

Figure 1, with its gentle modulations, creates the impression that the policy was enforced in a stable manner over time. Nothing could be further from the truth. Especially since the early 1980s, when the economic and political reforms disabled routine enforcement mechanisms, cadres have had to rely heavily on campaigns imposed from above to convince fellow villagers to contracept. The campaign nature of policy implementation can be read in Figure 2, which shows the proportion of women undergoing birth control operations of different types each year. If routine work wereofprimary importance, there would be gradual change in the number of operations over time. What the figure shows, instead, are large peaks in surgery followed by gullies and then more peaks. At the high point, 1983, one in eight village women underwent birth control surgery. At the low point, 1984, only one in 20 had an operation. The campaigns clearly emphasized sterilization in 1983 and 1987; in other years IUD insertion appears to have been the focus of cadre efforts.

14 american ethnologist
79 80 81 82 83 84 85 86 87
YEAR

Sterilization IUD Other

Figure 1. Percent ot women contracepting, by type, Shijiac~~n, 1979-87

resistance and negotiation: birth control in Shijiacun

In 1988, when I was in the village, resistance to the birth control policy was a highly elaborated practice. Economic means of defiance, which were manifold, have been described elsewhere (Greenhalgh 1993). Social strategiesZ0 were also common. For example, some couples whosesecondorthird children weredaughters adopted them outofthefamily, in hopes of being allowed another try for a son. Formal policy denounced this practice, explicitly excluding such couples from eligibility for further childbearing2' Another strategy was for parents with two daughters to move temporarily to a nearby city to have another child. While cadres and peasants alike denied the existence of temale infanticide, indirect evidence suggests that a few desperate parents may have resorted to letting their second or third baby daughter die in order to get permission to have another, hopefully male, child. There may have been other methods of concealing or disposing of unwanted daughters that Idid not find out about.22

controlling births and bodies 15
79 80 81 82 83 84 85 86 87
YEAR
Sterilization IUD Abortion

Figure 2. Percent of women undergoing birth control operations, by type, Shijiacun, 1979-87

Individual- and family-level actsexisted within, and gained forcefrom, a larger villagewide-- indeed, almost countrywid-nvironment of opposition to the harshest provisions of the policy. (Opposition was much weaker in the large cities.) In villages throughout the country, birth-planning cadres suffered withdrawal of political support, pressure from kinship networks, threats to their physical well-being, and other social injuries for failing to soften the

Because space restrictions preclude my detailing every oppositional strategy, I will focus on two of the most popular forms of resistance: the illicit removal of IUDs and failure to contracept.

IUD removal Unauthorized removal of IUDs has been a major problem for policy enforcers throughout the country (for the difficulties in Fujian, see Huang 1989:181). The Xianyang birth-planning regulations have consistently stipulated heavy penalties for those illegally removing IUDs, suggesting that the practice was far from rare in the area around Shijiacun.
16 american ethnologist

The data on contraceptive use in the village confirm this suspicion. In Shijiacun, 89 women had IUDs placed in their bodiesduringthe years 1979-87; 18 removed or expelled them within one, two, or three years. Fully one-third of the IUDs inserted during 1979-84 somehow came out within three years.

These IUD removals could have been violations of policy, unintended expulsions by women physically unable to carry the device, or licit actions authorized by birth cadres. Deciding which they were is made difficult by the fact that village women would have put themselves at some political risk by revealing that they had illegally removed their IUDs. They could, however, tell us whether the removals were authorized by local cadres; because none did, I assume none were.

Several types of evidence suggest that the majority of the 18 IUD losses were not physical expulsions but illicit removals by women wanting more children. First, clinical trials of the single-ring stainless steel IUD, the type most commonly used in China as a whole and in the area studied in the 19805, indicate that virtually all the physical expulsions occur within the first six months after insertion (Gao et al. 1986). In Shijiacun, only one-third of the IUD losses took place in the first six months.

Women's answers to the question of why their lUDs were removed provide a second clue to motivation (but only a clue, given the political constraints mentioned above). One IUD was taken out because its bearer became pregnant, while another was extracted when its wearer was sterilized. These were clearly legitimate removals. Four women told us straight out that they had had the devices removed becausethey wanted another child. These were just asclearly illicit extraction^.^^

The largest number of women-two-thirds of the total-replied that their IUDs were sponta- neously expelled (zidong toluo) or simply fell out (diaole). While Chinese statistics suggest that IUD expulsion rates are relatively high in that country,25 a look at the actions and circumstances of the villagers who somehow lost their IUDs suggests that the response "it fell out" may be a convenient gloss for "it was illegally removed."

To begin, the great majority (four-fifths) of women losing their IUDs within three years became pregnant a short time later. Fully 86 percent became pregnant within three months of having the IUD removed, implying a high level of sexual activity in a population not known for what demographers politely refer to as its "coital frequency." Such activity could well be motivated by a desire to conceive.

Women removing their IUDs also differed from women leaving the devices ir their bodies in ways that would be expected if their removals were strategies of resistance to the policy's provisions. The most obvious difference between IUD "resisters" and "acceptors" lies in the number of children they had at the time the device was inserted. Among women having lUDs implanted during 1979-84, the familiesof those removing them included 1.3 children, just half as many as the families of women leaving their IUDs in place. Two categories of women-those with only one child and those without a son-were particularly likely to have their lUDs taken out. Fully three-quarters of women in these circumstances who had an IUD inserted during 1979-84 promptly had it removed. Women who had no daughters were also strong candidates for removal: abouttwo-fifths ofsuch women who had IUDs inserted extracted them within three years. Thus, through IUD removal, women were making the point that they and their families badly wanted more than one child, a son, and, to a lesser yet still important extent, a daughter.

Zhao Cailing (a pseudonym) illustrates the lengths to which people went to have a second child. In March of 1984, she and her husband had their first child, a son. Following township regulations, Zhao had an IUD inserted three months later. But six months afterward, she had the IUD removed and began trying to get pregnant. She succeeded, but, in November of 1985, during a big birth-planning push in the township, her third trimester pregnancy was discovered

controlling births and bodies 17

and aborted. She and her husband managed to conceive a few months later and in January of 1987 had a baby girl.

Faced with such resolute resistance and lacking economic and administrative support, the birth-planning cadres had little choice but to negotiate new terms of policy enforcement. A close look at their responses to the manipulation of IUDs suggests the nature of the bargain struck: women who violated the reproductive norms of formal policy but had fewer children

. .

than the community minimum (twb children, at least one son) were allowed to carry their pregnancies to term. Those who tried to exceed community norms, however, were prevented from doing so. Thus, all but one of the women with subminimal families (that is, one son, one daughter, or two daughters) who removed their IUDs and became pregnant were allowed to carry the pregnancy to term. By contrast, every woman who already had the community ideal of one son and one daughter and became pregnant again was targeted for an abortion or sterilization. This rule of thumb fell particularly heavily on another villager, Cao Yumei, whose third pregnancy was aborted during the December 1986 campaign, one month before the scheduled delivery. There can be no doubt that this abortion was mandated by birth planning: Cao unhappily identified it as such.

The informal policy on IUDs thus was to tolerate removals, even if unauthorized, by those whosefamilies weretoo small by community standards, buttoget tough with those who already had their share of children. I would maintain that the community supported-indeed, was an accomplice26 in creating-this mode of policy enforcement. From constant propaganda in the media, in the marketplaceindeed, everywhere they went-the villagers knew that their neighbors who were also cadres had no choice but to enforce some kind of birth policy, and that they had to target someone during the birth control campaigns ordered by officials at higher administrative levels. Far from contesting the state's right to restrict their childbearing, the peasants accepted it and built it into their cultural and political assumptions about how to go about constructing their families. In a process of politically constrained community rationing of reproductive rights, the peasants evolved their own (state-influenced, as we have seen) norms of legitimate family size and, through "negotiations" that were both implicit and explicit, got the cadres to adopt them as enforcement standards.

noncontraception Throughout the one-child policy period, all married couples have been legally obligated to practice birth control. In practice, however, couples in certain situations have been temporarily exempt from this requirement. Those exempt include women pregnant with or waiting to conceive their first child, women in the immediate postpartum period, and women who are divorced, widowed, infertile, or menopausal. A national survey conducted in 1982 estimates that 18 percent of married women of reproductive age fall into these categories (Qiu et al. 1984). In Shijiacun, an average of 42 percent of married women of childbearing age were not contracepting during the years 1979-87. Assuming that the distribution of women by age and parity is roughly similar to that nationwide, then one-quarter (42 minus 18, or 24 percent) of the women should have been contracepting but were not. (Another 58 percent should have been contracepting and were.)

As we have seen, the Chinese state has attempted to legislate not only the practice, but also the type of contraception couples use. Since 1982, state policy has stipulated that all women with one child use an IUD and that one member of couples with two or more children undergo sterilization. To see how these rules fared in Shijiacun we look at contraceptive practice at the end of 1987.

As one might expect, not one childless couple was contracepting. (At no time did state regulations require those without children to protect themselves against pregnancy.) Only 15 percent of couples with onechild were protected against conception, the vast majority by means of the IUD. Contraceptive rates were much higher for couples with two or more children.

18 american ethnologist

Seventy-four percent of those with two children, 80 percent with three, and 82 percent with four or more were using some method to avert conception. But that method was rarely sterilization. Only 23 percent of couples with two children and 30 to 40 percent with three or more offspring were sterilized. These are rather high levels of noncompliance with state policy: fully 85 percent of couples with one child were violating the rules on contraceptive use, and some 60 to 75 percent of those with two or more were contravening the regulations on contraceptive type.

A closer look at who was breaking the rules suggests that failure to follow them was not mere oversight, but rather strategic behavior aimed at avoiding sterilization and ending up with more children than state policy allowed. Shown in Figure 3, patterns of contraceptive use by family size and composition correspond closely to what we would expect from our knowledge of the villagers' desires for two children-a son and, if possible, also a daughter. Thus, among couples with onechild, those with a son weretwiceas likely to becontracepting as thosewith a daughter (19 and 10 percent, respectively). Among parents with two children, those with oneor two boys were tour times more likely to be contracepting than those with two girls (79 and 86 percent, compared to 21 percent). While levels of contraceptive use were slightly higher among couples with two sons than among those with one of each sex, rates of sterilization were lower, suggesting a hope of eventually having a daughter.

Levels of contraception were uniformly high-roughly 75 to 85 percent-among couples with three children, but this is because there were virtually no parents with three daughters. Apparently those whose first three children were girls found a way to pare their families back to twoso as to have anotherchancefora boy. (As we will see below, couples with threechildren were easy targets for permanent contraceptive surgery during birth control campaigns.) Among three-child families, those with all sons had the highest levels of contraception but the lowest levelsof sterilization. Like two-son families, they might have been hoping to be able to try again for a daughter.

Cadres dealt with noncontraception in the same way they handled IUD removal: by negotiating new, informal norms that accommodated the most pressing peasant desires. Interviews with both cadres and peasants suggest that the cadres' rules of thumb were as follows. Couples with one child were urged to adopt contraception, but they were not unduly pressured to do so. History had shown that this was a waste of time, since the great majority of women with only one child having an IUD inserted lost no time in removing it to have another. Couples were allowed to have a family that was consistent with community norms on the understanding that once that ideal had been achieved they were expected to adopt long-term contraception. Empathizing with peasant fears of sterilization, cadres allowed that method to be an IUD. Some level of noncontraception was tolerated, even among couples with two and three children, as long as the couple did not make trouble. Once they created headaches-for example by trying to have, or, worse yet, succeeding in having, more children than regarded as necessary-they were punished by being targeted for an abortion or sterilization during the next campaign.

The tendency to target "greedy" couples emerges clearly from the evidence on abortion. Four-fifths of the 31 women undergoing abortion during 197947 already had families consid- ered ideal or near-ideal by community standards. Most had one son and one daughter, with the remainder having either two sons or one son and two daughters. As these family compositions would suggest, cadres targeting women for abortion paid due respect to their and their husbands' preference for sons: women undergoing abortion during 1979-87 had an average of

1.2 sons, but only .8 daughters. While the targeting process accommodated peasant fertility desires, the campaign nature of policy enforcement meant that a substantial number of thew abortions were performed later than the peasants or cadres would have preierred. Nearly half were carried out during the second or third trimester; the average duration of pregnancy at the time of termination was 4.1 months.

controlling births and bodies 19
Percent of Couples of Given Family Size and Composition Group

d

N L 0 (D 0

The ultimate contraceptive method, sterilization, was used with great care. Every one oithe 45 couples who was sterilized during the one-child period had at least the minimum family stipulated by community norms. The great majority (84 percent) had achieved the village ideal of one son and one daughter, while the rest had reached the village minimum of two sons. Waiving the state's rule that all couples with two children be sterilized, cadres allowed those whose first two were daughters to have another try for a son before being required to obtain permanent protection against pregnancy.

While any couple with the community ideal was liable to be targeted for sterilization during a campaign, two groups appear to have been particularly vulnerable: those who had escaped detection and had more children than the community standard, and those who had caused inordinate trouble, by, for example, having multiple abortions. About one of every six couples sterilized during these years had gotten away with having more children-usually more sons-than local norms allowed. One such egregious overreproducer, at least by local standards, was the couple Tian Chunlan and her husband, who somehow managed to have four children, two sons and two daughters, during the 1980s. This couple-which is to say Tian-was an easy target for sterilization in the December 1987 campaign. The tendency to target those requiring multiple abortions is difficult to demonstrate, given the small number of cases involved, but it is suggested by the fact that about one-tenth ofthe women sterilized during these years had had two or more abortions. Only 2percent of all village women had undergone two or more abortions. The most dogged of these women was Duan Jumei. By 1983 she already had two daughters and a son, yet still she and, presumably her husband, wanted another child. Refusing to contracept, she became pregnant three times, each resulting in an abortion (two late in the second trimester), before being decisively targeted for sterilization in the campaign of December 1987.

reproductive gains and their costs

By resisting state policy and forcing the birth cadres to negotiate local policy, women definitely improved their family-building prospects. Yet these demographic gains cost them dearly in terms of risk of bodily harm (see also Anagnost 1988). We look first atthe demographic results of local reproductive politics, then examine some paradoxes entailed by successful resistance.

negotiated family sizes From Table 1 we learn that Shijiacun couples by and large ended up with more children than formal policy allowed. The table includes all women married from 1971, the year the later-longer-fewer policy was put into effect. The oldest women represented here were in their early thirties when the one-child policy was announced. Although some had completed their childbearing by that time, many continued to give birth into the one-child policy period, and all were subject to that policy's rules on contraception and abortion. hly discussion focuses on those married between 1971 and 1983, most of whom had completed their childbearing by the end of 1987. Group averages for this subset of women can be found at the bottom of the table.

The table indicates that the great majority of women marrying between 1971 and 1983 had achieved the community ideal of two children by 1987. Seventy percent had two children, while another 22 percent had three. Despite the one-child-withexceptions policy officially in effect since 1979, a mere 7 percent had only one child at the end of 1987.

The data on family composition provide additional evidence of the peasants' success in promoting their reproductive agenda. Sixty-four percent of those marrying between 1971 and 1983 ended up with the village ideal of at least one son and one daughter (column [k] plus column [I]). Another 22 percent had two or three sons, considered an acceptable alternative to

controlling births and bodies 21

Table 1. Reproductive outcomes ot Shijiacun women by year of marriage, year-end 1987
Nurnber of Children         -.         Family Composition     
                    More    Nurnber
Year of                         No             2-3D     2-3s         than     of
marriage     0     1     2     3     4+     children     10     1S     NOS     NOD     lD,lS     1 D+l S     women
        (b)    (c) (Percent of total)     (d)     (el         (g)     (h)     (i)(Percent oi total)     (j)     (k)         
1971-73         .O     40.7     59.3     .O                                 
1974-76         .O     73.1     23.1     3.8                                 
197 7-79         10.0     75.0     15.0     .O                                 
1980-81         9.1     75.8     15.2     .O                                 
1982-83         1 1 .9     78.6     7.1     2.4                                 
1984-85         53.3     40.0     3.3     .O                                 
1986-87         57.4     2.1     .o     .o                                 
All Years         23.6     51.6     15.1     .9                                 
1971-83                             
Notes: D = Daughter and S = Son.                             
Source: Calculated from reproductive histories.                             

one of each sex. Seven percent had two or three daughters (in all cases but one, it was two daughters), while another 7 percent had only one son or one daughter. While 14 percent of couples with less than the village minimum may seem to imply a high level of reproductive dissatisfaction, we saw in Figure 3 that the vast majority of people in these situations were not contracepting. One suspects they were biding their time, waiting to conceive again.

These large reductions in the time spent in pregnancy and on child care have affected every aspect of women's lives, opening up options never before imaginable. While this is not the place to assess how their lives might differ from those of their mothers and older sisters, I want to note that in the late 1980s the women of Shijiacun felt this freedom from reproductive obligations as personally liberating. Although their economic opportunities remained limited to family horticulture, they loved the open air of the fields and cherished the opportunity to make important contributions to the family economy. They may have hated the way birth planning was enforced, but they were far from unhappy with its effects on other aspects of their lives.

the costs of resistance: bodily risks Resistance had paradoxical effects for women: while lifting the limit on the number of children they could have, at the same time it put their bodies at risk. For example, the more children a woman succeeded in having, the more likely she was to be labeled a troublemaker and targeted for surgery. The more women contested the implantation of IUDs, the greater the number of insertions, extractions, and reinsertions of the device they had to undergo. Getting pregnant outside the community ideal always entailed the risk of detection and subsequent "mobilized" (i.e., coerced), often late-term, abortion. While first trimester abortions are relatively safe (in China such operations are performed by vacuum aspiration), later abortions pose elevated health riskseven where medical facilities are excellent, a situation that does not always obtain in China.

Table 2 contains some evidence of the cumulative effects of the struggle over birth control on women's bodies. We see here that these costs were borne disproportionately by the older women, who had been in the reproductive combat the longest. Among those married during the 1970s, the vast majority had undergone one or more birth control operations by 1987. Between 19 and 45 percent had been sterilized. The highest rates of sterilization-36 to 45 percent-were found among those who had the misfortune to enter marital and reproductive lifeduring 1977-81, when the later-longer-fewer policy was abandonedforthe one-child policy and the sterilization and other big campaigns of the early one-child period raged through the political system.

The data on IUD insertion and abortion also suggest a certain degree of bodily risk. The probability of having undergone at least one IUD insertion rose steadily from 20 percent for women married during 1984-85 to 85 percent for those married during 1971-73, suggesting that IUD use was an almost inevitable phase in the contraceptive lifespan of Shijiacun women. Roughly oneof eight women married in the 1970s and early 1980s had had to undergo multiple IUD insertions. About one ofevery four villagewomen married in the 1970s had had an abortion by 1987, and one in eight had undergone the trauma of a second or third trimester abortion.
conclusion

China's birth control program brought about what may be the most rapid medicalization of human reproduction in history. In less than a quarter of a century, the proportion of reproduc- tive-age women contracepting leapt from near zero to, if not the nearly 80 percent officially claimed (China Population Information and Research Center 1990), then at least 50 or 60, or even 70 percent. Because contraception entailed bodily intervention, the majority of women became subject to control by the state and its army of political and medical cadres.

controlling births and bodies 23

In Shijiacun and probably thousands of villages like it, women were not only victims of the birth control policy, although they certainly were that. They were also agents, who, acting for themselves and their families, actively contested policy elements they did not like, forcing local cadres to negotiate the terms of policy implementation. Through their actions, villagers protested many of the formal rules, compelling negotiations over such issues as who would have to undergo surgery and how many children of what sex they would be allowed to raise. The result was the creation of informal rules partly accommodating their desires. Local policy deviated significantly from national and provincial policy, a pattern that obtained elsewhere in the country as well (Kaufman et al. 1989; Potter and Potter 1990).

While strategies of resistance altered local policy in ways women sought, their effects were contradictory. Along with a degree of reproductive freedom and an increase in the time available for activities outside the reproductive domain, opposition to state policy brought the potential for corporal harm. The more women contested the policy, the greater the risks to their bodies from tubal ligations, repeated IUD insertions and extractions, frequent abortions, late-term abortions, and botched operations performed in haste during mass birth control campaigns.

Resistance to the policy also reinforced women's social subordination. By pressing so hard for sons, women succeeded in biasing the policy toward males. Not only informal village policies, but also formal provincial and national policies were modified to accommodate peasant demands for sons. Acting as agents for the patriarchal family and themselves, women etched their own inferiority into the birth policy, with worrying consequences for the future.

While peasant families were able to challenge state domination of the reproductive domain, ironically, they managed to reproduce state control over their childbearing even as they opposed it. Peasant ideas about the good family bore the clear stamp of state-defined norms, and peasant-cadre negotiations assumed the state's right to micromanage the process of family construction. Thus, in reproduction, as in other areasof social and cultural behavior (Siu 1989), state control remained awesome years after the state retreated from direct control of village life.

The China material adds layers of new meaning to the family demographers' view of the family as a political construction. In the villages studied, the family was nothing but a political construction, in the sense that its size and composition were virtually dictated by the rules of the one-child policy. But if the family was a political construction, the policy was also a family construction in that family resistance to its provisions constrained local agents of the state to modify it to accommodate peasant demands.

I began this article by suggesting that the study of third world reproduction, which is now largely the study of family demography, would benefit from a broadening to include the issues of state intervention, resistance to control, and bodily risks addressed by the feminist and political-economic literatures. In China, of course, one cannot help but consider these issues, because the state has structured the domain of family formation in such a way that one is unable to study childbearing withouttaking statecontrol, individual resistance, and demographic issues into account. But even where the state does not force our intellectual hand, bringing the feminist and politicaleconomic issues to bear on reproduction would help broaden our arena of inquiry. At more microlevels, a feminist perspective would bring into focus the body as a site of struggle; strategies of manipulation, acceptance, and rejection of reproductive technologies; and trade- offs people make between the goals of bodily control, health and safety, and ideally sized families. At more macrolevels, a political-economic inquiry would draw our attention to the state, the medical profession, and international agencies as shapers of the means and ends of fertility control; to strategies of resistance and accommodation to state interventions; and to the consequences of these strategies for state policies and programs. By combining these three perspectives into one, we could create a larger anthropology of reproduction that might finally

controlling births and bodies 25

pull this fundamental issue of human existence from the periphery to the center of anthropo- logical inquiry.
notes

Acknowledgments. This article has beneiited greatly from the critical comments of Caroline Bledsoe, Judith Bruce, Kay Johnson, Li Nan, Vivienne Shue, Wei Jin, and three anonymous reviewers. Earlier discussions with Bina Agarwal, Judith Bruce, Mary Ann Burriss, Joan Kauiman, Barbara Mensch, and Wei Jin were also very helpful. The field research was supported by the National Science Foundation (Grant #BNS-861812); funding ior data analysis was provided by the National lnstituteior Child Health and Human Development (Grant #1 R01 HD 28689-01). The article is based on hand tabulations oi a small part of a very large data set.

    "First wave" American and European feminism developed in the latter part of the 19th century and remained active through the 1920s.

    A major outcome of this autocritique by organizations supporting iamily planning has been the development oi the "quality of care" approach to iamily planning (Bruce 1990; Mensch 1992). Marking what outside observers consider a "positive and decisive break with the past" (Hartmann 1987:143), the quality oi care standpoint maintains that greater responsiveness to women's health and safety concerns and respect ior women's right to reproductive seli-determination are important both in their own right and because they promote more sustained contraceptive use and thus iertility decline.

    1 am grateful to Judith Bruce ior illuminating discussion of this point.

    1 thank Bina Agarwal ior her thoughts on this issue.

    Of course, the theoretical and ideological roots oi demography proper are quite different (Hodgson 1991). There are also divisions within (cultural) demographic anthropology, with some closer to demogra- phy, sociology, andlor sociobiology, and others closer to the anthropological mainstream as it has developed, say, in the pages oi this journal. The latter iorm a distinct majority.

    Chineseresearchers also iaced risks from overstepping the boundsoipolitical propriety. WhileChinese authorities realized long ago that resistance to the one-child policy was undermining their ambitious population control program, they prefer that Chinese scholars emphasize behaviors that are positive and policy enhancing. Discussions with top ofiicials in the population sector give me confidence that this article will be viewed as constructive in China because it suggests concrete ways the one-child policy can be improved.

    The types oi conditions under which couples were allowed to have a second child were iairly restrictive. For example, couples in which both spouses are returned overseas Chinese or in which one is a handicapped veteran were allowed two children.

    Article 49 of the 1982 Constitution states that "both husband and wife have the duty to practice iamily planning" (Fifth National People's Congress 1982).

    In its first two years, the policy sought to curtail childbearing through both late marriage and birth control. When the 1980 Marriage Law reduced the efiective marriage age, officials responsible ior birth planning were iorced to drop the emphasis on late marriage, focusing most oi their efforts on birth control within marriage. The policy sought to encourage (rather than mandate) late marriage, late childbearing, and long spacing. In rural areas, however, these policy stipulations were extremely difficult to enforce, and thus in practice were given little stress.

    From Xianyang City campaign document oi February 1983 (in author's possession) and Weinan Commune document of December 1982 (based on notes taken during field research).

    The exceptions to these requirements include those for whom the operations are medically unadvis- able, older women, and those using other birth control methods and having no contraceptive iailure ior more than eight years. The 1986 regulations continued to advocate sterilization ior couples with two children, but mandated the operation for those with threeor more.

    During 1979 and 1987, over three-quarters oisterilization operations conducted in China as a whole were periormed on women (Ministry of Public Health 1988:79).

    For more on the role Chinese social scientists in the formation and eniorcement of the birth policy see Greenhalgh 1990b.

    Provincial and national income figures come from State Statistical Bureau 1988:826.

    In the 1980s, most sonsdivided from theirfamilieswithin a yearor twoaiter marriage. Inextraordinary cases, the sons partitioned before marriage.

    Discussions with older women suggest that attitudes toward birth control were quite different in the 19605, when modern contraception was iirst introduced in the village. At that time many women, especially those who had suffered miscarriages and seen children die in infancy, actively welcomed the technological opportunity to be free from fourth, iifth, and higher-order pregnancies.

    The research of Kauiman and her colleagues suggests that rural Chinese women have good reason for concern about the medical repercussions of contraceptive use (Kaufman et al. 1992). Their 1987 iield study of four counties in Heilongjiang and Fujian Provinces revealed women to be poorly iniormed about the risks associated with use oi diiierent birth control methods. Only one-quarter of women experiencing side eifects from IUD use knew in advance to expect them. In one county, one-iifth of the women who were

26 american ethnologist

sterilized were not told the operation was permanent. Few township or village birth-planning cadres they interviewed could identiiy contraindications to IUD use; those iamiliar with the IUD's side effects had no knowledge of how to treat them. Isolated press reports hint that complications from sterilization operations may be common problems (see, ior example, Cao 1985).

    Commonly referred to as commune reiorm, these political changes involved establishing a township government at the commune level and turning the commune into an economic organization. Brigades becamevillagers' committees, and production teams were transiormed into villagers' small groups. The aim oi these changes was to reduce party penetration oi the economy. For details see Shue 1984.

    The oiiicial data on contraception are subject to methodological problems at iour crucial levels: use oi leading questions by survey investigators (Banister 1987:179), misreporting by individuals, ialsification by local cadres, and use oi misleading (and unexplained) indicators by government statisticians. For example, the denominator might excludecouples who had permission to conceive, producing higher-than- normal contraceptive use rates. Actors at each of these levels have a strong motivation to bias contraceptive use rates upward. As virtually all demographic data irom large-scale surveys are gathered by government employees, independent checks that might detect such problems are absent.

    By strategy here I mean any action designed to aifect the way the policy was enforced. By definition all such strategies were reactive. While some took the iorm oiioot dragging, state pressure to comply with the norms of the policy ensured that even these apparently passive iorms oi behavior took on an active, oppositional character.

    My data on adoption include only iormal adoptions, in which all rights to the children are legally relinquished. Evidence irom elsewhere suggests that informal adoption practices, in which children are "lent" to a relative elsewhere for a iew years, have become common during the one-child policy period. A recent study suggests that as many as one-half of the "missing girls" in demographic surveys of the 1980s may have been adopted out into other families, the majority by iniormal means Uohansson and Nygren 1991).

    Such methods may include abandonment and sale, which have become common in rural China in recent years. During a return visit to the villages in 1993, 1 learned of the use oi prenatal sex determination by ultrasound, iollowed by abortion of femaleietuses. These latter practices appear to bemuch less common in Shaanxi than in the more developed eastern provinces (Zeng et al. 1993).

    1 am grateful to Kay Johnson ior discussion on this point.

    Judging from the timing oi the IUD extractions and second pregnancies, none oi these women was authorized to have a second birth. Authorized second births are supposed to take place four or more years aiter the iirst.

    Clinical trials suggest an expulsion rate oi 12.6 percent in the first year (Cao et al. 1986). Only three of the women saying their IUDs "iell out" were under 25, an age group noted ior its high rates oi IUD failure (Sivin and Schmidt 1987).

    The word is Siu's, irom the title of her 1989 book.

references cited

Anagnost, Ann Stasia 1988 Family Violence and Magical Violence: The Woman as Victim in China's One-Child Family Policy. Women and Language 1 1 :16-22. Ardener, Shirley 1975 Introduction. In Perceiving Women. Shirley Ardener, ed. Pp. vii-xxiii. London: Malaby. Banister, Judith 1987 China's Changing Population. Staniord, CA: Stanford University Press. Berger, Peter L., and Thomas Luckmann 1966 The Social Construction of Reality: A Treatise in the Sociology oi Knowledge. New York: Doubleday. Bledsoe, Caroline

1990 The Politics of Children: Fosterage and the Social Management of Fertility among the Mende oi Sierra Leone. In Births and Power: Social Change and the Politics of Reproduction. W. Penn Handwerker, ed. Pp. 81-1 00. Boulder, CO: Westview.

Bledsoe, Caroline, and Uche Isuigo-Abanihe

1989 Strategiesof Child-Fosterageamong Mende Crannies in Sierra Leone. In Reproduction and Social Organization in Sub-Saharan Africa. Ron J. Lesthaeghe, ed. Pp. 442474. Berkeley: University of California Press.

Browner, C. H., and Sondra T. Perdue 1988 Women's Secrets: Bases ior Reproductive and Social Autonomy in a Mexican Community. American Ethnologist 15:84-97. Bruce, Judith 1987 Users' Perspectives on Contraceptive Technology and Delivery Systems: Highlighting Some Feminist Issues. Technology in Society 9:359-383. 1990 Fundamental Elements of the Quality oi Care: A Simple Framework. Studies in Family Planning 21:61-91.

controlling births and bodies 27

Bruce, Judith, and S. Bruce Schearer 1983 Contraceptives and Developing Countries: The Role oi Barrier Methods. Paper presented at the International Symposium on the Regulation of Human Fertility, Stockholm. Burns, John P. . . 1985-86 Local Cadre Accommodation to the 'Responsibility System' in Rural China. Pacific Affairs 58:607-625. , , Carter, Anthony T.

In press Agency and Fertility: For an Ethnography oipractice. In Anthropology and Fertility: CuItureIHis- tory1GenderlPower in Demographic Liie. Susan Greenhalgh, ed. Cambridge: Cambridge University Press.

Chen Pi-chao, and Adrienne Kols 1982 Population and Birth Planning in the People's Republicof China. Population Reports, Series 1, 25. Pp. 1577-61 8. China Population Information and Research Center 1990 China Population Newsletter 7, no. 4. Beijing. Coale, Ansley 1. 1984 Rapid Population Change in China, 1952-1982. Washington, DC: National Research Council, National Academy Press. Dixon-Mueller, Ruth, and Judith Wasserheit 1991 The Culture of Silence: Reproductive Tract Infections among Women in the Third World. New York: International Women's Health Coalition. Donzelot, Jacques 1979 The Policing oi Families. New York: Pantheon. Ehrenreich, Barbara, and Deirdre English 1978 For Her Own Good: 150 Years of the Experts' Advice on Women. Garden City, NJ: Anchor. Fiith National People's Congress 1982 The Constitution of the People's RepublicoiChina, promulgated on December 4, 1982. Beijing: Foreign Languages Press. Foucault, Michel 1978 The History of Sexuality, 1. New York: Pantheon. Fricke, Tom In Dress Political Economv and Fertility: History, Culture, and Demographic Events in a Central ~imala~an in Demographic Life.

Village. ln~nihro~olog~ and ~ertilit~:'~ulture/~istor~/Gende~/~bwerSusan Greenhalgh, ed. Cambridge: Cambridge University Press. Gal, Susan

1991 Between Speech and Silence: The Problematics oi Research on Language and Gender. In Gender at the Crossroads oi Knowledge: Feminist Anthropology in the Postmodern Era. Micaela di Leonardo, ed. Pp. 175-203. Berkeley: University of Caliiornia Press.

Gao Ji, Shen He, Zheng Su, Fan Hui-min, Wu Ming-hui, Han Li-hui, and Yao Guong-zhen 1986 A Randomized Comparative Clinical Evaluation of the Steel R~ng, the VCu200 and the TCu220c IUDs. Contraception 33:443-454. Gao Junfeng

1985 Ying Zhongshi Jieza Funu de Huli Gongzuo (We Should Pay Attention to the Treatment of Sterilized Women). Jiankang Bao,Jihua Shengyu Ban (Health Gazette, Birth Control Edition [Beijingl), December 5.

Giddens, Anthony 1979 Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis. Cambridge: Cambridge University Press. Ginsburg, ~a~e 1989 Contested Lives: The Abortion Debate in an American Community. Berkeley: University oi California Press. Ginsburg, Faye, and Rayna Rapp 1991 The Politics of Reproduction. Annual Review oi Anthropology 20:31 1-343. Ginsburg, Faye, and Amy Lowenhaupt Tsing 1990 Introduction. In Uncertain Terms: Negotiating Gender in American Culture. Faye Ginsburg and Amy Lowenhaupt Tsing, eds. Pp. 1-16. Boston: Beacon. Gordon, Linda 1977 Women's Body, Women's Right: A Social History oi Birth Control in America. Harmondsworth, Middlesex, England: Penguin. Greenhalgh, Susan 1986 Shifts in China's Population Policy, 1984-86: Views irom the Central, Provincial, and Local Levels. Populat~on and Development Review 12:491-51 5. 1990a Toward a Political Economy of Fertility: Anthropological Contributions. Population and Devel- opment Review 1 6:85-106. 1990b Population Studies in China: Privileged Past, Anxious Future. Australian Journal of Chinese Aiiairs 24:357-384. 1990c The Evolution oithe One-Child Policy in Shaanxi, 1979-88. TheChina Quarterly 122:191-229.

28 american ethnologist

1993 The Peasantization oi the One-Child Policy in Shaanxi. In Chinese Families in the Post-Mao Era. Deborah Davis and Stevan Harrell, eds. Pp. 21 9-250. Berkeley: University of Caliiornia Press. Hartmann, Betsy 1987 Reproductive Rights and Wrongs: The Global Politics of Population Control and Contraceptive Choice. New York: Harper and Row. Hodgson, Dennis 1991 The Ideological Origins of the Population Association of America. Population and Development Review 17:l-34. Huang Shu-min 1989 The Spiral Road: Change in a Chinese Village through the Eyes oi a Communist Party Leader. Boulder, CO: Westview.

1990 Folk Reproductive Medicine in North China: The Cultural Constructs of Pregnancy, Delivery and Post-Natal Care. Paper presented at the Third International Congress on Traditional Asian Medicine, Bombay.

Johansson, Sten, and Ola Nygren 1991 TheMissing Girls of China: A New Demographic Account. Population and Development Review 17:35-51. Kauiman, Joan, Zhang Zhirong, Qiao Xinjian, and Zhang Yang 1989 Family Planning Policy and Praaice in China: A Study of Four Rural Counties. Population and Development Review 15:707-729. 1992 The Quality of Family Planning Services in Rural China. Studies in Family Planning 23:73-84. Kertzer, David I., and Dennis P. Hogan 1989 Family, Political Economy, and Demographic Change: TheTransformation of Liie in Casalecchio, Italy, 1861-1921. Madison: University of Wisconsin Press. Launay, Robert

In press Illegitimacy in an African Muslim Community. In Anthropology and Fertility: Culture/His- torylGender1Power in Demographic Liie. Susan Greenhalgh, ed. Cambridge: Cambridge University Press.

Levine, Nancy E. 1987 Differential Child Care in Three Tibetan Commun~ties: Beyond Son Preierence. Population and Development Review 13:281-304. Liu Zheng 1981 Population Planning and Demographic Theory. In China's Population: Problems and Prospeas. Liu Zheng et al., eds. Pp. 1-24. Beijing: New World Press.

Martin, Emily 1987 The Woman in the Body: A Cultural Analysis of Reproduction. Boston: Beacon. 1990 The Ideology of Reproduction: The Reproduction oi Ideology. In Uncertain Terms. F. Ginsburg

and A. L. Tsing, eds. Pp. 300-31 4. Boston: Beacon. 1991 The Egg and the Sperm: How Science has Constructed a Romance Based on Stereotypical Male-Female Roles. Signs 16:485-501. Mensch, Barbara 1992 Quality of Care: A Neglected Dimension. In The Health ofwomen: A Global Perspective. Marge Koblinsky, Judith Timyan, and Jill Gay, eds. Pp. 235-253. Boulder, CO: Westview. Ministry oi Public Health 1988 Zhongguo Weisheng Tongji Tiyao 1987 (Chinese Health Statistical Digest 1987). Beijing. Oakley, Ann 1984 The Captured Womb: A History of the Medical CareoiPregnant Women. Oxford: Basil Blackwell.

-

Oi, Jean C. 1989 State and Peasant in Contemporarv China: The Political Economv oi Villaee Government.
Ll

Berkeley: University oi Caliiornia ~iess. ' Parish, William L., and Martin King Whyte 1978 Village and Family in Contemporary China. Chicago: University oi Chicago Press. Potter, Sulamith Heins, and Jack M. Potter 1990 China's Peasants: The Anthropology oi a Revolution. Cambridge: Cambridge University Press. Qiu Shuhua, Wu Shutao, and Wang Meizeng

1984 A Survey oi China's Birth Control Among Women of Child-Bearing Age. In Analysis on China's National One-per-Thousand-Population Fertility Sampling Survey. Pp. 139-1 44. Beijing: China Popu- lation lniormation Center.

Rapp, Rayna 1990 Constructing Amniocentesis: Maternal and Medical Discourses. In Uncertain Terms. F. Ginsburg and A. L. Tsing, ed. Pp. 2842. Boston: Beacon.

1991 Moral Pioneers: Women, Men, and Fetuses on a Frontier oi Reproductive Technology. In Gender at the Crossroads oi Knowledge: Feminist Anthropology in the Postmodern Era. Micaela di Leonardo, ed. Pp. 383-395. Berkeley: University oi California Press.

Rosaldo, Renato 1989 Culture and Truth: The Remaking of Social Analysis. Boston: Beacon. Roseberry, William 1988 Political Economy. Annual Review oi Anthropology 17:161-185.
controlling births and bodies 29

1989 Anthropologies and Histories: Essays in Culture, History, and Political Economy. New Brunswick: Rutgers University Press. Ross, John A., W. Parker Mauldin, Steven R. Green, and Romana Cooke 1992 Family Planningand Child Survival Programs as Assessed in 1991. New York: Population Council. Rowland, Robyn 1987 Technology and Reproduction: Reproductive Choice Reconsidered. Signs 12:512-528. Schneider, Jane, and Peter Schneider 1984 Demographic Transitions in a Sicilian Rural Town. Journal oi Family History 9:245-273. Shaanxi Provincial Bureau oi Statistics 1987 Shaanxi Tongji Nianjian, 1987 (Shaanxi Statistical Yearbook, 1987). Xi'an. Shue, Vivienne 1988 The Reach oi the State: Sketches oi the Chinese Body Politic. Stanford, CA: Staniord University Press. 1984 The Fateof the Commune. Modern China 10:259-283. Siu, Helen F. 1989 Agents and Viaims in South China: Accomplices in Rural Revolution. New Haven, CT: Yale University Press. Sivin, Irving, and Frederick Schmidt 1987 Efiectiveness oi IUDs: A Review. Contraception 36:55-84. Song Jian 1981 Population Development-Coals and Plans. In China's Population: Problems and Prospects. Liu Zheng et al., eds. Pp. 25-31. Beijing: New World Press. Stanworth, Michelle, ed. 1987 Reproductive Technologies: Gender, Motherhood and Medicine. Minneapolis: University oi Minnesota Press. State Statistical Bureau 1988 Zhongguo Tongji Nianjian, 1988 (Statistical Yearbook oi China, 1988). Beijing. Unger, Jonathan 1985-86 The Decolle~ivization oi the Chinese Countryside: A Survey oiTwenty-Eight Villages. Paciiic Affairs 58:585-606. Warwick, Donald P. 1982 Bitter Pills: Population Policies and Their Implementation in Eight Developing Countries. Cam- bridge: Cambridge University Press. Wasserstrom, Jeffrey 1984 Resistance to the One-Child Family. Modern China 10:345-374. White, Tyrene 1992 The Origins oi China's Birth Planning Program: A Revisionist View. Paper presented at the Conierence on Engendering China, Harvard University, Cambridge, MA. Whyte, Martin King, and S. Z. Gu 1987 Popular Response to China's Fertility Transition. Population and Development Review 13:471-

493.

Wolf, Arthur P., and Chieh-shan Huang 1980 Marriage and Adoption in China, 1845-1 945. Staniord, CA: Stanford University Press.

-

Wolf, Eric R. 1982 Euro~eand the People Without History. Berkeley: Un~versity oi Californ~a Press. Zeng Y i 1989 Is the Chinese Family Planning Progaram 'Tightening Up'? Population and Development Review 15:333-337. Zeng Yi, Tu Ping, Cu Baochang, Xu Yi, Li Bohua, and Li Yongping 1993 Causes and Implications of the Recent Increase in the Reported Sex Ratio at Birth in China. Population and Development Review 19:283-302.

submitted March 17, 1992 accepted May 25, 1992

30 american ethnologist

Comments
  • Recommend Us